methyl and total mercury spatial and temporal …...2001). as a result, the fish, water, and...

Methyl and Total Mercury Spatial and Temporal Trends in

Surficial Sediments of the San Francisco Bay-Delta

Assessment of Ecological and Human Health Impacts of Mercury in the Bay-Delta Watershed

CALFED Bay-Delta Mercury Project Final Report

Submitted to:

Mark Stephenson

California Department of Fish and Game Moss Landing Marine Labs

7544 Sandholdt Road Moss Landing, CA 95039

Submitted by:

Wesley A. Heim

Moss Landing Marine Laboratories 8272 Moss Landing Rd

Moss Landing, CA 95039 [email protected] (email)

831-771-4459 (voice)

Dr. Kenneth Coale Moss Landing Marine Laboratories

8272 Moss Landing Rd Moss Landing, CA 95039

Mark Stephenson

California Department of Fish and Game Moss Landing Marine Labs

7544 Sandholdt Road Moss Landing, CA 95039

CCAALLFFEEDD MMeerrccuurryy PPrroojjeecctt -- MMLLMMLL FFiinnaall RReeppoorrtt

EXECUTIVE SUMMARY Recent studies indicate significant amounts of mercury are transported into the Bay-Delta from the Coastal and Sierra mountain ranges. In response to mercury contamination of the Bay-Delta and potential risks to humans, health advisories have been posted in the estuary, recommending no consumption of large striped bass and limited consumption of other sport fish. The major objective of the CALFED Bay-Delta Mercury Project “Assessment of Ecological and Human Health Impacts of Mercury in the Bay-Delta Watershed” is to reduce mercury levels in fish tissue to levels that do not pose a health threat to humans or wildlife. This report summarizes the accomplishments of the Moss Landing Marine Laboratories (MLML) and California Department of Fish and Game (CDF&G) at Moss Landing as participants in the CALFED Bay-Delta Mercury Project. Specific objectives of MLML and CDF&G include: 1. Determination of the spatial distribution of total mercury and methyl mercury in surficial sediments of the Bay-Delta. Identify locations within the Bay-Delta having high mercury methylation potentials (as indicated by the methyl mercury to total mercury ratio) (Task 4A1 & 4A2). 2. Determination of the temporal changes in total mercury, methyl mercury, and the methyl mercury to total mercury ratio at six locations within the Bay-Delta (Task 4A1 & 4A2). 3. Determination of the mercury methylation potentials in Coastal and Sierra Range river and lake sediments (Task 4A1 & 4A2). 4. Determination of the mercury methylation potential of wetland areas within the Delta (Task 4A1 & 4A2 Phase 2). The initial survey of Bay-Delta sediments was conducted October, 1999. Samples were collected for the determination of total mercury, methyl mercury, percent Loss on Ignition (LOI), and grain size, as well as water column measurements of temperature and pH. Sediment samples were collected monthly from May, 2000 – November, 2001 at six locations in the Bay-Delta. Samples were analyzed for total and methyl mercury, and LOI. Sediment collections were made in October, 2001 from the Coastal Range and Sierra Range for the determination of mercury methylation potentials. Transects across three wetlands were conducted May 2001. Sediments were collected from the interior, middle, exterior, and control site of each wetland area for the determination of total mercury, methyl mercury, and LOI. All raw data is given in Appendix B. Major Findings (Working Hypotheses):

• In the Bay-Delta, total mercury concentrations decreased north to south and increased east to west while methyl mercury concentrations increased north to south. The highest methyl mercury sediment concentrations were found in the


central Delta. The central Delta had the greatest mercury methylation potential when compared to the surrounding tributaries.

• Solid phase methyl mercury concentrations varied seasonally with highest

concentrations occurring during late spring/summer.

• Coastal mountain streams had two orders of magnitude higher total mercury concentrations than Sierra mountain streams and the methyl mercury concentrations were equivalent. Sediments from lakes of the Coastal Range and Sierra Range were equivalent in total mercury and methyl mercury.

• Inner areas of wetlands consistently had higher solid phase methyl mercury

concentrations and methyl mercury to total mercury ratios when compared to outer areas of wetlands and the control site.


INTRODUCTION

The CALFED Mercury Project – Background and Objectives

Background

Between 1850 and 1980 California was the nation’s leading producer of mercury,

contributing about 100 million kilograms to the world market (Churchill 2000). Most of

the mercury was mined in the coast range of California. Early mercury processing using

furnace volatilization of the cinnabar ore and cold surface condensation of elemental

mercury was inefficient and approximately 35 million additional kilos of the metal are

estimated to have been released into the Coastal Range watersheds (Churchill 2000). The

mercury mined in the coast range was shipped across the central valley to the Sierra

Nevada where it was used in gold mining activities. Six million kilos of mercury are

thought to have been lost in placer and lode gold mining in the Sierra Nevada (Churchill

2000). Recent studies have determined that large amounts of mercury are still being

transported annually into the San Francisco Bay-Delta from both the Coastal Range and

the Sierra Nevada (Larry Walker and Associates 1997, Foe and Croyle 1998, Roth et al.

2001). As a result, the fish, water, and sediment of the central valley and Bay-Delta have

elevated levels of mercury.

The primary human health concern of environmental mercury contamination is

exposure through the consumption of contaminated fish. The dominant form of mercury

in fish is the organic species methyl mercury (Bloom 1992). Methyl mercury is formed

principally by sulfate-reducing bacteria in surficial sediments (Gilmour et al. 1998).

Methyl mercury is readily bioaccumulated by aquatic organisms and biomagnified in the

food web. Methyl mercury is a potent human neurotoxin, with developing fetuses and

small children being most at risk at extremely low levels (White et al. 1995). In response

to the mercury contamination of the Bay-Delta and the potential risks to humans, health

advisories and interim health advisories have been posted in the estuary, recommending

no consumption of large striped bass and limited consumption of other sport fish (Office

of Environmental Health Hazard Assessment 1994, San Francisco Bay Regional Water

Quality Control Board 1995). Elevated concentrations of mercury in fish tissue may also


represent a hazard to piscivorous wildlife. Species most at risk are fish-eating birds and

mammals.

Site Description

The San Francisco Bay and Sacramento-San Joaquin Delta (Bay-Delta) is the

largest estuary on the west coast of the United States. The dominant source of fresh

water to the Bay-Delta is the Sacramento River with smaller contributions supplied by

the San Joaquin, Mokelumne, and Cosumnes Rivers. The Cosumnes River is the only

river which flows into the Bay-Delta unobstructed by dams. The Bay-Delta receives the

runoff from 40 percent (60,000 sq miles) of California’s land. The Bay-Delta covers

738,000 acres with hundreds of miles of waterways within its boundaries. These

waterways, originally formed from the natural meandering of rivers through the

marshland, are now held in place with banks made up of rip-rap armor and extensive

dikes. Historic seasonal wetlands, created by the annual flooding as a result of rivers

overflowing natural levees, have all but disappeared from the Bay-Delta. Large acreages

of marsh habitat are now found at wildlife recreation areas and restoration sites such as

the Cosumnes River Preserve (northeast delta), the Lower Sherman Island Wildlife Area

(west delta), and the Yolo Bypass Wildlife Area (northwest delta). Historically, seasonal

wetlands and marshes were the predominant habitat features in the central delta. Today,

the majority of these seasonal wetlands have been reclaimed for agricultural uses

(Nichols et al. 1986).

Much of the marshes and wetlands in the delta are constrained to the fringes of

channels along rip-rap banks and narrow islands in the center of waterways (Picture 1).

The subsidence of reclaimed swamp land as a result of oxidation and mechanical

compaction has left much of the Bay-Delta below sea level necessitating the more than

1,000 miles of levees for protection against flooding (Picture 2).

The Bay-Delta is critical to the state of California. The Bay-Delta system is not

only important as the fresh water source for 22 million Californians, but also for the crops

grown in the rich delta soils as well as the habitat it provides for many fish and wildlife

species.


Objectives

The present mercury project is funded by CALFED, an ambitious long-term

program to improve water management in the Bay-Delta Estuary. The program includes,

in addition to water management, ecosystem restoration, levee stabilization, and water

quality components. The main goal of the mercury element of the water quality program

is to reduce mercury concentrations in fish tissue to levels that do not pose a threat to

wildlife or human health.

Organization

The CALFED Mercury Research Group consists of 15 principal investigators

from State and Federal agencies, universities, private analytical laboratories, and non-

profit agencies (see: http://loer.tamug.tamu.edu/calfed). The mercury research group is

headed by Mark Stephenson of the California Department of Fish and Game (CDF&G) at

Moss Landing Marine Laboratories (MLML)


MLML and CDF&G Project Objectives

MLML and CDF&G were involved with four tasks as part of the overall CALFED

Mercury Research Project. A description of the tasks is briefly outlined below.

1) Spatial Patterns of Methyl Mercury Production (Task 4A1 & 4A2).

a. Determine the spatial distribution of total mercury and methyl mercury in surficial sediments of the Bay-Delta.

b. Use the methyl mercury to total mercury ratio to identify locations of high methylation potential.

c. Relate spatial patterns to distributions of total mercury and percent loss on ignition (LOI)(used as a proxy for total organic carbon).

2) Temporal Patterns of Methyl Mercury Production (Task 4A1 & 4A2).

Determine the temporal changes in methyl mercury to total mercury ratios in surficial sediments at six locations within the Bay-Delta.

3) Mercury Bioavailability (Task 4A1 & 4A2 phase 2).

Compare the methylation potential of Coastal and Sierra Range sediments.

4) Methylation Efficiency of Delta Wetlands (Task 4A1 & 4A2 phase 2).

Determine the difference in methylation potential between wetland and non-wetland areas within the Delta.

METHODS

Sampling for Initial Survey (Determination of Spatial Patterns, Objective 1a-1c)

A broad scale survey of the Bay-Delta system was conducted October 10, 1999

through December 1, 1999. Ninety-six sites within the Bay-Delta (Figure 1) and one

hundred fifty sites from South Bay to San Pablo Bay and the Delta tributaries were

sampled within a two-month period with the majority of the Delta samples collected

within the first five days. Sampling locations in South Bay and San Pablo Bay are shown

in Figures 2-4. Water temperatures in the Bay-Delta at the time of sampling averaged

15 ºC, and the Sacramento River was flowing at 10,000-19,000 cfs.


A geographical information system (GIS) was used to generate random sampling

locations within each habitat type defined by the Wetland Inventories base map of the

Bay-Delta . Once in the field, many locations were changed due to problems with access,

as a result the sampling was semi-random. Table 1 gives a list of the habitat types and

number of samples collected within each habitat during the Delta survey.

Sediment samples were collected remotely using the “Sludge O Matic” (SOM), a

sampler designed and built at Moss Landing Marine Labs described below and in

Appendix A. Samples were collected for the analysis of total mercury, methyl mercury,

Loss on Ignition (LOI), and grain size (percent fines = < 63 µm).

Seasonal Sampling (Determination of Temporal Patterns, Objective 2)

Six sites within the Bay-Delta system were selected for monthly monitoring

(Figure 5). The locations were selected to represent a variety of habitat types from

different areas of the Bay-Delta. A position and brief description of the habitat type for

each location is given in Table 2. Sampling frequency was monthly beginning May 2000

and increased to twice-monthly June 2001. Stations were located with a hand held GPS

to assure re-sampling to within a few meters. Sediment was collected using the SOM

sampler. The samples were analyzed for total mercury, methyl mercury, and LOI. Grain

size (percent fines as described above) analysis was conducted on a selected sub-set of

the samples.

Bioavailability Sampling (Objective 3)

Sediment collections were made during October, 2001 in the coastal mountain

range (Cache Creek and other areas) and in the Sierra Nevada range (Cosumnes River

and other areas)(Figure 6). Two seasonal lakes (Capay Dam and Cache Creek Settling

Basin) were also sampled to be able to make a comparison between the permanent lakes

and seasonal lakes of the coastal range. Samples were collected using a scoop to remove

the top 1 cm of sediments from depositional areas in the streambeds. A scoop was used

rather than the SOM as the sediment type was predominately coarse grained sand which

prevents the shutter door on the sampler from sealing properly. Lake samples were

collected using the 0-1 cm interval of sectioned cores. Both sieved (60µm) and unsieved


samples were collected. Sieved samples were collected to prevent bias introduced by

differences in grain size from confounding results. Sieves were made from 60µm plastic

mesh. The methyl to total mercury ratio was used as a proxy for methylation efficiency

of the sediments; higher methyl mercury concentrations found in the sediments per unit

of total mercury indicate higher methylation potential and bioavailability.

Wetland Sampling (Objective 4)

Investigation of the methylation efficiency of wetland versus non-wetland areas

were conducted during May, 2001. Name, location, and a brief habitat description of the

three wetland areas studied are listed in Table 3. At each area studied, sampling was

conducted along a transect, one station furthest into the interior of the wetland (inner),

one midway from the inner station to the outside of the wetland (middle), and one just

outside of the wetland (outer). The Mandeville Island station had sediment containing

peat, channels greater than two meters deep, and the water was well mixed and visually

appeared to be equivalent to the adjacent San Joaquin River. Webber Point and Fourteen

Mile Slough had silty sediment free of peat, limited water circulation, and were close to

agricultural areas. At each station two replicate cores were collected, the sediment from

each core was extruded and the sediment sub-sampled at the interval of 0-2 cm. The

control sites were located on the San Joaquin River away from the wetlands and the

sediments were silty and similar in appearance to Webber Point and Fourteen Mile

Slough sediments. Samples were analyzed for total mercury, methyl mercury, and LOI.

Sediment Sampler

The upper portions of the sediment column are thought to be the most important

in terms of trophic transfer and sediment/water flux. It is here that the sediment contacts

the water, and the transition from oxic to anoxic usually occurs within a few millimeters

in delta sediments (Gill, personal communication). The maximal rates of Hg methylation

have been shown to occur just below the oxic/anoxic transition zone in sediments

(Gilmour et al. 1992). In addition it is from this sedimentary layer that most diffusional

flux originates (Gill et al. 1999), and it is here where most benthic infaunal invertebrates

live. Therefore, a decision was made to sample the suficial 0-0.5 cm sediment interval


for the “survey” and “seasonal” work. This interval does not accurately characterize the

total amount of mercury locked up in the sediment reservoir of the Bay-Delta, but does

target the sediment stratum from which mercury is most likely to be mobilized into the

water column. However, there are no commercially available samplers capable of

reliably sampling this portion of the sediment column. A sampler was therefore designed

and built to cleanly capture the top 0.5 centimeters of sediment. Picture 3 shows the

SOM fully assembled with the trap door in the closed position and the trigger line

attached ready for remote use. Picture 4 shows a close-up of SOM with the trap door

open exposing the cleanly captured top 0.5 cm of sediment. Appendix A contains

detailed instructions for the construction of the SOM sampler. The primary benefits of

this sampler were the large sediment yield per deployment (~ 135 g), the straight forward

sub-sampling of the targeted surface sediment interval, and the consistency in sampling

the surface sediment interval. Unlike sectioning cores for a targeted sediment interval,

where the sub-sampling is tedious and time consuming, this sampler required nothing

more than scooping the sediment out of the sampler and into the sample container. The

sampler is deployed remotely from a small boat by attaching a hand pole or line and

weight depending on water column depth. In addition, the sampler may be used directly

by a SCUBA diver.

Clean sample collection

Sediment samples were collected into 60 mL wide-mouth borosilicate glass jars,

with Teflon™ lined polyethylene caps (I-CHEM™ EPA-clean, or equivalent acid-

cleaned) using standard clean techniques. Clean techniques briefly described are as

follows: sampling crews consisted of at least two persons, one wearing fresh clean room

gloves “clean hands” and the other holding the sampler “dirty hands” (Picture 5). The

clean hands person was responsible for transferring sediment from the sampler into the

sample jar. The dirty hands person assisted while not coming into direct contact with the

sample until the sample was contained. The jars were filled to 60-80 % of the container

capacity with sediment (to minimize head-space losses of volatile Hg, while allowing for

expansion due to freezing). The samples were kept on dry ice in the field and during

transport back to the lab. Samples were frozen at < -10˚C (ordinary freezer at lowest


setting) with a maximum holding time of 1 year. All homogenization and other handling

of low-level samples (< 100 ng/g) were performed by clean-room gloved personnel in an

area known to be low in atmospheric Hg. It should be noted that the clean techniques

employed here were developed in conjunction with the MLML trace metals laboratories

and Gary Gill of Texas A&M University at Galveston. These methods have been shown

to be non-contaminating for mercury and methyl mercury at low environmental levels.

Laboratory Analysis

All mercury and methyl mercury analysis was performed using methods and

quality assurance/quality control detailed in the CALFED Mercury Project QAPP

(Puckett and van Buuren 2000, http://loer.tamug.tamu.edu/calfed/). Analysis of LOI was

performed following standard method ASTM D2974. Grain size analysis was performed

using standard methods of Plumb (1981).

RESULTS AND DISCUSSION

Quality Assurance/Quality Control

MLML participated in three intercomparison studies and sent splits of sediment

samples to Frontier for total and methyl mercury analysis. All QA results are listed in

Appendix C of this report. Results of the intercomparison studies can be obtained at the

CALFED Mercury Project website (http://loer.tamug.tamu.edu/calfed/DraftReports.htm).

The results of the first intercomparison study showed good interlaboratory precision for

total mercury in sediment. Results for methyl mercury in sediment showed a larger

degree of variability possibly due to a slight low bias by one of the participating

laboratories (van Buuren, 2003). The third intercomparison study showed good

interlaboratory precision for both total and methyl mercury in sediments (van Buuren,

2002). The relative percent differences (RPDs) between split sediment samples analyzed

for total mercury at MLML and Frontier ranged from 1 – 20 percent, while the RPDs for

split sediment samples analyzed for methyl mercury ranged from 10 – 172 percent. The

total mercury split data shows good interlaboratory precision. The high RPDs of samples


analyzed for methyl mercury may be attributed to the difficulties associated with splitting

a heterogeneous natural sediment sample containing decomposing vegetation and woody

debris (a common characteristic of delta sediments) rather than a potential analytical

problem. It should be noted that six sieved samples were split and analyzed for methyl

mercury by both MLML and Battelle Laboratories resulting in RPDs ranging from 0.8 –

56.1 percent. The lower RPDs for sieved samples than unsieved samples points toward

the splitting of heterogeneous sediment samples as a potential problem which may

confound results of split samples far more than any analytical variation between

laboratories. In addition, Bloom (2003), found thawed sediment samples kept

refrigerated resulted in an increase in methyl mercury concentration when compared to

samples kept frozen prior to analysis. Further investigations should be made to set up

protocols for split samples to minimize variation from sample handling protocols. For

future sediment split intercomparison studies, strict protocols should be followed to

insure the samples are split in the field and kept frozen until analyzed. At no time should

intercomparison samples be thawed and analyzed at one laboratory with an aliquot

refrozen and transported to a second laboratory for analysis. Currently it remains unclear

if the observed differences in methyl mercury concentrations measured at participating

laboratories are attributable to analytical procedures, splitting methodology, or natural

variation.

Small Scale Variation

An estimate of the small scale variation in sediment concentrations of methyl

mercury and total mercury at locations sampled in the Bay-Delta is listed in Tables 4 & 5.

The relative percent difference (RPD) of duplicate methyl mercury samples collected

during the initial survey ranged from 19 to 108 percent (Table 4), while RPD’s for total

mercury were 4 to 82 percent (Table 5).

Figure 7 shows the small scale variation of methyl mercury in sediments collected

from the west and east side of the river channel at East Columbia Cut in the central delta.

The small scale variation between replicates collected from the west side of the channel

was less than that of samples collected from the east side of the channel. The sediment of

the west side was fine grained and homogenous with no submerged aquatic vegetation


visible in the sample (Picture 4) while the sediment of the east side was a heterogeneous

mix of sand and silt covered with both submerged and floating aquatic vegetation

(Picture 6). These results emphasize that small scale variation in methyl mercury

concentration is expected to increase in locations within the Bay-Delta with

heterogeneous sediment types having an abundance of submerged aquatic vegetation

present.

Geographical Trends

San Pablo Bay and South Bay sediment total mercury concentrations were

typically found to be around 0.3 ppm (Figure 2). The total mercury concentrations in

sediments around the Oakland inner harbor were approximately 1 ppm with Brooklyn

harbor (Station # 283) being ten times higher at 9.4 ppm (Figure 2). Methyl mercury

sediment concentrations in San Pablo Bay were less than 1 ppb (Figure 3). A few

locations in the middle South Bay around Oakland Harbor had methyl mercury

concentrations above 2 ppb (Figure 3). The highest methyl mercury concentration found

in the South Bay was 15.89 ppb at Coyote Hill Slough (Station # 310) (Figure 3). There

was no correlation between methyl mercury and LOI. The amount of methyl mercury to

the total mercury present in sediments of San Pablo Bay and South Bay was

approximately one half to one percent (Figure 4). Two locations, Richmond Inner

Harbor (Station # 273) and Coyote Hill Slough (Station # 310) had percent methyl

mercury values of three and eight respectively (Figure 4). The methyl mercury

concentrations in Oakland harbor appear to be a result of higher total mercury

concentrations and not the result of a favorable habitat for mercury methylation as

indicated by the ratio of methyl mercury to total mercury being consistent with

surrounding locations sampled. In contrast, the high methyl mercury sediment

concentrations in the Richmond Inner Harbor (Station # 273) and Coyote Hill Slough

(Station # 310) are likely due to favorable conditions for mercury methylation and

methylation is occurring independent of the total amount of mercury present in the

sediment.

Total mercury in the Bay-Delta decreased moving north to south from the

northern tributaries of Prospect Slough and Cosumnes River into the central and southern


Delta (Figure 8). Total mercury concentrations of sediments in the northern tributaries

were as high as 0.4-0.5 ppm in contrast to the central Delta with mercury concentrations

from 0.1-0.3 ppm. The southern Delta sediments typically had very low mercury

concentration; the greatest cluster of samples below the method detection limit (MDL)

were found in the southern portion of the Delta. These results are consistent with the

findings of Slotten et al. (2003).

A second spatial trend observed was an increase in total mercury (0.15 to 0.30

ppm) moving east to west from the central Delta into Suison Bay and Grizzly Bay

(Figures 8). Total mercury concentration increased with an increase in longitude when

constrained by 38.00-38.15 degrees latitude (Figure 9). The Carquinez Straight, Suison

Bay, and Grizzly Bay areas had a large number of sites with high total mercury

concentrations. Generally, sediments in the San Pablo and Suison Bays averaged 0.3

ppm with some sites above 0.5 ppm total mercury. It is possible for sediment

concentrations to be biased by changes in grain size. In this case, however, the increase

in total mercury concentration moving out of the Delta was not a bias introduced by a

change in grain size as the normalized and un-normalized total mercury concentration

versus longitude yield similar correlation coefficients (normalized to percent fines data

not shown).

In contrast to the geographic patterns of total mercury, methyl mercury generally

increased from north to south moving into the central Delta from the northern tributaries

(Prospect Slough and Cosumnes River) (Figure10). The central Delta consistently had

higher methyl mercury concentrations than the perimeter waterways and adjacent bays.

Methyl mercury concentrations were very low or non-detectable (MDL= 0.019 ppb) in

the Sacramento River and San Joaquin River channels moving west, out of the central

Delta. At the convergence of the Sacramento and San Joaquin Rivers (Sherman Island

area) methyl mercury concentrations increased to values as high as many of the central

Delta sites (Figure 10). Joyce Island in Grizzly Bay had the highest methyl mercury

sediment concentration (9.3 ppb) of all sites sampled in the North Bay and Delta.

The ratio of methyl mercury to total mercury was used to identify locations within

the Bay-Delta with high mercury methylation potential (Figure 11). The central Delta

had the greatest potential for mercury methylation; many sites in the central Delta had


greater than 2 % of the total mercury as methyl mercury. Franks Tract had the largest

measured ratio (2.6 %). In addition, other sites within the central Delta appeared to have

large methylation potentials as total mercury concentrations were below the MDL and a

considerable amount of methyl mercury was present (Figure 11- grey scale bars).

However, the ratios at these sites were estimated using the MDL value of 10.5 ppb as a

value for total mercury and, therefore, must be considered only as estimated potentials.

All of the tributaries surrounding the central Delta (e.g. Prospect Slough, Cosumnes

River, Sherman Island, the upper San Joaquin River, and southern Delta) had very low

mercury methylation potentials. Although Honker Bay had many sites with relatively

high methyl mercury concentrations only one site (Joyce Island) showed a large potential

for mercury methylation (2.3 % methyl mercury). A comparison of the methylation

potential of the central Delta and the surrounding tributaries indicates the central Delta

possesses environmental conditions favorable for mercury methylation. An

understanding of the factor/s forming these optimal conditions is desirable to control the

production of methyl mercury in the Bay-Delta sediments.

Numerous field-based studies have suggested that organic carbon is positively

correlated with methyl mercury in sediments (Choi and Bartha 1994, Hurley et al. 1998,

Krabbenhoft et al. 1999). In addition, lab based studies have shown that organic carbon

is positively correlated to methyl mercury production (Olson and Cooper 1976, Furutani

and Rudd 1980, Wright and Hamilton 1982, Lee and Hultberg 1990). Consistent with

these studies, LOI (a proxy for total organic carbon) was significantly correlated (R2 =

0.32) to methyl mercury concentration in Delta sediments (Figures 12). However, the

low correlation coefficient indicates other factors, in addition to LOI, likely control

methyl mercury production in Delta sediments.

Regnell et al. (1997) indicated that total mercury correlated positively with methyl

mercury in surficial sediments of a seasonally stratified lake in southern Sweden. One of

the reactants in the mercury methylation reaction is Hg2+ therefore the concentration of

Hg2+ in a system might be expected to have an effect on the rate of formation, and the

concentration, of methyl mercury (Kelly et al. 1995). Rudd et al. (1983) reported a linear

relationship between amount of Hg2+ added to lake sediments and rate of formation of

methyl mercury, demonstrating that in a system where all else remained constant,


methylation was first order with mercury concentration. In this study, a weak positive

correlation was found between methyl mercury and total mercury concentrations in

sediments (Pearson’s r2= 0.19, n = 99, p<0.01) (Figure 13). The correlation between

methyl mercury and total mercury indicates total mercury may play a role in the

production of methyl mercury, however, the correlation is weak and is dependent on

throwing out 4 % of the data as outliers.

In a study of the Florida Everglades, factors such as wetland density, low surface

water pH, and sulfide were shown to be important in controlling methyl mercury

production (Gilmour et al. 1998). Factors other than total mercury and organic carbon

may have a much stronger influence on the production of methyl mercury in the Bay-

Delta. Strategies to control methyl mercury production in the Bay-Delta should consider

the relative importance of total mercury, organic carbon, as well as the many factors

mentioned above. Currently, the relative contribution to which each of these factors

influences methyl mercury production in the Bay-Delta is unknown.

Trends across Habitat Types

All samples collected within the delta during the initial survey fell into habitat

types defined by wetland inventories GIS maps. The methyl mercury concentrations for

all of the samples assigned to a particular habitat type were averaged to give an average

methyl mercury concentration and error for each habitat type. Table 1 lists the methyl

mercury sediment concentrations of each habitat sampled in the delta during the initial

survey. The marsh habitats had methyl mercury concentrations above 1 ppb. Lakes and

ponds also had concentrations above 1 ppb, however, it must be pointed out that only two

samples were collected from this habitat type. Farmed wetlands, seasonal wetlands, open

water, and mudflats had methyl mercury concentrations less than 1 ppb. A large number

of open water samples were collected and as a result there is high confidence that this

habitat type is accurately characterized. Ideally all habitat types should be sampled as

thoroughly as the open water habitats were. The riparian woodland and upland habitats

had methyl mercury sediment concentrations less than 0.25 ppb. This work identifies the

marsh habitat type as a potential source of methyl mercury as the average methyl mercury

concentrations are higher than all other habitat types sampled. In addition, the farmed


wetlands could be an important source of methyl mercury as the percent coverage of this

habitat type is large with respect to the total area of the delta and the methyl mercury

concentrations are relatively high compared to other habitat types.

Seasonal Trends

Solid phase methyl mercury showed distinct seasonal patterns at locations within

the central Delta and Cosumnes River (Figure 14, panels c-f). Methyl mercury

concentrations at Franks Tract, Connection Slough, and Cosumnes River were elevated

twice per year (Figure 14, panels c, d, f). The largest peak occurred during summer and

the smaller peak occurred over winter. The winter peak, although lesser in magnitude

(approx. 2 ppb), was generally a longer lasting feature (3-4 months) while the summer

peak reached a maximum (approx. 6 ppb) and decreased within 1-2 months. Methyl

mercury in the sediments at Prospect Slough remained constant (approx. 1 ppb) over the

length of the study (Figure 14, panel a). Sherman Island had an extended period of

elevated solid phase methyl mercury concentration summer and fall of 2000, followed by

a decrease in concentration throughout the remainder of the study (Figure 14, panel b).

Connection Slough was the only site sampled where a repeating summer methyl

mercury peak was observed. No summer peak was found at Franks Tract or White

Slough during the summer of 2000 (Figure 14, panels c, e), perhaps as a result of the

sampling frequency. The sampling frequency was once per month at Franks Tract and

White Slough during the summer of 2000, and it is probable that this interval allowed the

summer peaks to go undetected. The 2001 summer peak at Franks Tract occurred very

rapidly, and sampling twice per month was necessary to capture the event. Gill et al.

(1999) also reported a rapid seasonal increase and decrease of methyl mercury, which

necessitated frequent sampling during a study done in Lavaca Bay, Texas.

Total mercury sediment concentrations showed no seasonal trends (Figure 14,

panels a-f). However, at Sherman Island, total mercury concentrations were higher

during the first half of the study and generally decreased throughout the second half of

the study; a similar trend was found for methyl mercury (Figure 14, panel b). Total

mercury concentrations demonstrated increased variability at Prospect Slough and

Sherman Island, as compared with the central Delta and Cosumnes River. The largely


variable concentrations at Prospect Slough and Sherman Island may be the result of bias

due to mean grain size, as high water flow and a scoured bottom characterize these

locations. Within the central Delta, White Slough had the least variable total mercury

concentrations, and Cosumnes River consistently had the highest total mercury

concentrations (Figure 14, panel e, f).

Bloom et al. (1999) used the methyl mercury to total mercury ratio in sediments

to evaluate the seasonal variation of mercury methylation. Gilmour et al. (1998) found in

situ production of methyl mercury controlled concentration in the Florida Everglades.

The seasonal changes in sediment methyl mercury concentration observed in the Bay-

Delta were likely the result of increased in situ production resulting from a stimulation of

the microbial activity within the within the sampled stratum. Evidence of in situ

production is found in the methyl mercury to total mercury ratio being driven by changes

in methyl mercury meaning the methyl mercury concentrations were increasing as the

total amount of mercury remained unchanged. Pore water concentration gradients

determined by Gill (2003) also support in situ production as the source of methyl mercury

as the methyl mercury concentrations in pore waters were always higher than the

overlying water concentrations. Conversely, the spring and summer peak in methyl

mercury may have resulted from movement of the peak methylating layer into and out of

the sampled sediment interval and not from increased in situ methylation within the

sampled stratum. This could occur because the peak microbial production of methyl

mercury occurs in a thin zone near the oxic/anoxic interface in the vertical sediment

profile (Gilmour et al. 1992) and the oxic/anoxic boundary can be expected to shoal in

sediments during periods of low flow and high temperatures. However, dissolved oxygen

profiles determined on intact cores using microelectrodes indicate that the oxic/anoxic

interface remains in the top 0-1 cm interval in delta sediments (Gill, 2003). Furthermore,

solid phase methyl mercury vertical profiles in sediment cores collected from the delta

during this study do not support a vertical movement of methyl mercury with season that

would explain the observed temporal shifts in surficial methyl mercury concentrations

(Gill, 2003).

Sulfate reducing bacteria are especially important methylators of mercury

(Gilmour et al. 1992). Factors that increase sulfate reduction rates, such as high water


temperature, high availability of organic carbon, and high sulfate concentrations are

likely to increase the production of methyl mercury (Compeau and Bartha 1985, Gilmour

et al. 1992). Water temperature in the Delta fluctuates annually following a sinusoidal

curve with maximum summer water temperatures of approximately 20 ºC and winter

minimum temperatures of approximately 10 ºC. Consistent with the findings of Gilmour

et al. (1998) in a study of the Florida Everglades, methyl mercury sediment

concentrations in the Bay-Delta were elevated during periods of warm water

temperatures. In contrast to the positive correlation between LOI and methyl mercury

described in 0bjective 1c the seasonal increase in methyl mercury concentration occurred

independent of changes in LOI (data not shown). A possible explanation is the inability

of the LOI determinations to make any statement to the quality of the carbon present.

Stimulation of microbial activity and any resultant production of methyl mercury is

dependent in part on the availability of labile carbon; LOI is a description of the total

amount of carbon, both labile and refractory. Sediments containing a large fraction of

refractory material such as peat would give a high value for LOI with little benefit to the

microbial community.

There was no relationship between total mercury and methyl mercury

concentrations at the seasonal study sites which exhibited a summer increase in methyl

mercury production (Figure 15). The highest methyl mercury concentrations occurred

over a broad range of total mercury concentrations. In contrast to the overall picture,

however, if only the Sherman Island site is considered, there was a positive linear

relationship between methyl mercury and total mercury (R2 = 0.40) (Figure 15). A

predictive model of methyl mercury based on total mercury concentrations determined

from the Sherman Island data would not be useful on a basin wide scale as changes in

environmental conditions negated the first order relationship between total mercury and

methyl mercury.

Determining the primary factors responsible for the seasonal increase in

production of methyl mercury was beyond the scope of this study, although it is likely

that temperature plays a role, as does available organic carbon. The use of total mercury

concentration as a predictor of methyl mercury production was useful only in locations

where seasonal changes in environmental conditions were minimal. Further research is


needed to understand the processes and combination of environmental factors that

promote mercury methylation during spring and summer in the Bay-Delta.

Mercury Bioavailability

During the California gold rush era, mercury was mined in the coastal mountains

and transported to the Sierra Mountain Range for use in Gold mining. Mercury mined

from the coastal range was in the form of cinnabar while the Sierra Range was

contaminated exclusively with refined mercury (elemental). These forms of mercury

have very different behaviors in terms of reactivity and solubility. One important

question to ask is are the mercury species found in these mountain ranges equally

available for mercury methylation (bioavailable). Finding dissimilar methyl mercury to

total mercury ratios in the coastal and Sierra range watersheds would be evidence that

mercury in the coastal and Sierra ranges are not equally available for methylation.

The comparisons of coastal versus Sierra range total mercury concentrations,

methyl mercury concentrations, and the methyl mercury to total mercury ratios for

unsieved sediment samples are shown in Figure 16. Coastal and Sierra range lakes had

equivalent levels of total mercury, methyl mercury, and methyl mercury to total mercury

ratios. Total mercury concentrations were much higher (18,000 ppb) in coastal mountain

streams, near mine sites, than Sierra mountain streams (49 ppb) and slightly higher in

coastal valley streams (122 ppb) than Sierra valley streams (33 ppb). Methyl mercury

concentrations were equivalent in coastal and Sierra mountain and valley streams. The

methyl mercury to total mercury ratios in Sierra mountain and valley streams were

significantly higher than coastal mountain streams. However, it is important to point out

the large standard error associated with the Sierra valley stream data.

In an attempt to reduce any bias introduced by grain size variability and determine

the methylation potential of only the fine-grained fraction, samples were sieved through a

60µm screen. Figure 17 shows the comparisons between coastal and Sierra range total

mercury concentrations, methyl mercury concentrations, and the methyl mercury to total

mercury ratios for sieved sediment samples. Trends in total mercury and methyl mercury

of sieved samples are similar to unsieved samples (Figures 16 & 17). Both coastal and

Sierra range lakes had equivalent total mercury, methyl mercury, and methyl mercury to


total mercury ratios (Figure 17). Total mercury concentrations were higher in coastal

mountain and valley streams than Sierra mountain and valley streams while methyl

mercury concentrations were equivalent. The methyl mercury to total mercury ratio in

Sierra mountain streams was significantly higher than coastal mountain streams but,

unlike the unsieved results no difference was found in the methyl mercury to total

mercury ratios of coastal and Sierra valley streams.

The permanent lakes were significantly lower (Least Significant Difference

multiple comparison test, p<0.05) in methyl mercury concentration and the methyl

mercury to total mercury ratio than valley or mountain streams (Figures 17). This is in

contrast to the seasonal lakes (Capay Dam and Cache Creek Settling Basin), in which

some of the highest concentrations of methyl mercury occurred (8 and 4 ng/g dry weight

sediment respectively). These results are consistent with other studies showing newly

flooded wetlands and reservoirs experience an increase in methyl mercury production

(Bodaly et al. 1993, Kelly et al. 1995, Kelly et al. 1997). Kelly et al. (1997) concluded

that the large increases in methyl mercury concentrations were a result of increased net

production of methyl mercury rather than release of methyl mercury that was in the

wetland prior to flooding. The increase in methyl mercury, at these newly flooded areas,

is likely linked to increased microbial activity in response to a change in environmental

conditions. Three changes in environmental conditions known to stimulate methylation

of mercury are 1) sudden death of vegetation supplying a large amount of organic carbon

to become available for decomposition, 2) high decomposition leading to an increase in

anaerobic habitat, and 3) mercury methylation stimulated by increased temperature

(Kelly et al. 1997). Capay Dam and the Cache Creek Settling Basin experience all three

of these changes in environmental conditions on an annual basis allowing an increase in

methyl mercury production to occur independent of an increase in total mercury.

The most significant finding of the mercury bioavailability study was the higher

methylation potentials (greater bioavailability) of sediments from the Sierra mountain

streams compared to coastal mountain streams. This is perhaps a problem of definition

because the total mercury concentrations near the mine site stations in the coast Range

are relatively high, greater than 10,000 ppb, which drives the low methyl mercury to total

mercury ratios. However, the methyl mercury concentrations in the two watersheds were


equivalent, indicating that the large pool of mercury in the coastal range was largely

unavailable for methylation.

Current work suggests mercury in the upper watershed of the coastal range is

significantly less available for methylation than the mercury in the Sierra range.

However, as coastal range mercury is transported through the watershed into the Bay-

Delta it becomes more available for methylation. Other work supporting this working

hypothesis is that the sediment, clams, and fish collected at the mouths of the tributaries

coming from both ranges have equivalent total and methyl mercury concentrations (Davis

et al., 2003; Slotten et al., 2003; Foe et al., 2003).

Mercury at the mine sites is likely to start out as cinnabar and, through diagenic

processes in the sediment, becomes more bioavailable for methylation. Paquette and

Heltz (1995) showed that high concentrations of sulfide leading to the formation of

soluble polysulfide complexes increase the solubility of cinnabar. In addition, it has been

demonstrated that organic rich sediments are capable of enhancing the dissolution of

cinnabar (Ravichandran et al. 1998, Wallschlager et al. 1998). The Cache Creek

watershed has high levels of sulfide as a result of thermal springs (Domalgalski, 2003),

which could enhance cinnabar dissolution. The average LOI for Cache Creek is 6

percent making it unlikely that organic matter is responsible for the cinnabar dissolution.

Clearly, additional studies directed towards further testing of this working hypothesis are

needed. One question left unanswered is what role if any does differing environmental

conditions play in the methylation efficiency of coastal versus Sierra range derived

mercury.

Wetland Study

The investigation of wetlands to determine if methylation potentials were higher

in the interior than adjacent waterways showed clear patterns in both methyl mercury

concentrations and the methyl mercury to total mercury ratios (Figure 18 & 19). Methyl

mercury concentrations at the interior of all wetland areas studied were higher than

concentrations at the exterior of the wetlands (Figure 18). In addition, the methyl

mercury to total mercury ratio was highest at the interior of all three wetlands studied


(Figure 19). LOI was significantly correlated to both methyl mercury (p< 0.001) and the

methyl mercury to total mercury ratio (p< 0.001) (Data not shown).

Numerous studies have shown wetland areas to be areas of high methyl mercury

production (St. Louis et al. 1994, Hurley et al. 1995, Rudd 1995, St. Louis et al. 1995,

Krabbenhoft et al. 1999). The common features of these areas are limited circulation,

proximity to agricultural areas, and lack of peat in the sediments. The LOI levels in the

sediment in the interior of these wetlands appeared to be high (7-50%) relative to those in

the main, well flushed, part of the Delta (usually <5%). Water circulation in the interior

of the wetland areas was limited and led to increased water temperature and residence

time. The large amount of organic material present and the limited circulation of the

water likely led to hypoxic water conditions. The combination of these environmental

conditions set the ideal stage for mercury methylation in the presence of reactive mercury

(Kelly et al. 1997). Clearly the wetland areas within the Bay-Delta have a high potential

for production of methyl mercury. Further research is needed to quantify the production

of methyl mercury in the wetland habitats and determine the contribution these habitats

make with respect to the mercury budget of the entire Bay-Delta.

CONCLUSIONS In brief summary, the major findings of the study are listed below.

• The highest methyl mercury sediment concentrations were found in the central

Delta. The central Delta had the greatest mercury methylation potential when compared to the surrounding tributaries.

• Solid phase methyl mercury concentrations varied seasonally with highest

concentrations occurring during late spring/summer.

• Coastal mountain streams had two orders of magnitude higher total mercury concentrations than Sierra mountain streams and the methyl mercury concentrations were equivalent.

• Interiors of wetlands consistently had higher solid phase methyl mercury

concentrations and methyl mercury to total mercury ratios than fringes of wetlands.


ACKNOWLEDGEMENTS This work would not have been possible without the field assistance and/or analytical work of Sean Mundell, Bettina Sohst, Lisa Berrios, Susan Von Thun, Myah Gunn, John Haskins, Elizabeth Sassone, Susan Coale, Autumn Bonnema, Amy Byington, Tam Voss, and the MPSL clean lab crew. In addition, Dylan Service was largely responsible for the flawless operation of the many small boats used to sample the Bay-Delta.Special thanks to the reviewers Dr C. Gilmour, Dr. D. Krabbenhoft,Dr. J. Wiener, and Dr. J. Rudd for all of their good advice and recommendations.


LITERATURE CITED Bloom, N. S. 1992. On the chemical form of mercury in edible fish and marine

invertebrate tissue. Can. J. Fish. Aquat. Sci. 49:1010-1017. Bloom, N. S., G. A. Gill, S. Cappellino, C. Dobbs, L. Mcshea, C. Driscoll, R. Mason, and

J. Rudd. 1999. Speciation and cycling of mercury in Lavaca, Bay, Texas, sediments. Environ. Sci. & Technol. 33:7-13.

Bloom, N. S., 2003. Task 7C In CALFED Final Report titled "An Assessment of Human Health and Ecological Impacts of Mercury in the Bay-Delta Watershed."

Bodaly, R. A., J. W. M. Rudd, and R. J. P. Fudge. 1993. Mercury concentrations in fish related to size of remote Canadian shield lakes. Can. J. Fish. Aquat. Sci. 50:980-987.

Choi, S. C., and R. Bartha. 1994. Environmental factors affecting mercury methylation in estuarine sediments. Bull. Environ Contam. Toxicol. 53:805-812.

Churchill, R. 2000. Contributions of Mercury to California's Environment from Mercury and Gold Mining Activities- Insights from the Historical Record. in Assessing and Managing Mercury from Historic and Current Mining Activities, San Francisco, Ca, USA.

Compeau, G. C., and R. Bartha. 1985. Sulfate-reducing bacteria: Principal methylators of mercury in anoxic estuarine sediment. Appl. Environ. Microbiol. 50:498-502.

Davis, J. A., B. K. Greenfield, G. Ichikawa, and M. Stephenson. 2003. Mercury in Sport Fish from the Delta Region. Task 2A In CALFED Final Report titled "An Assessment of Human Health and Ecological Impacts of Mercury in the Bay-Delta Watershed."

Domagalski, J. L., C. N. Alpers, D. G. Slotten, T. H. Suchanek, and S. H. Ayers. 2003. Mercury and Methylmercury Concentrations and Loads within the Cache Creek Watershed, California, January 2000 through May 2001. In CALFED Final Report titled "An Assessment of Human Health and Ecological Impacts of Mercury in the Bay-Delta Watershed."

Foe, C., and W. Croyle. 1998. Mercury Concentrations and Loads from the Sacramento River and from Cache Creek to the Sacramento-San Joaquin Delta Estuary. Staff Report Central Valley Regional Water Quality Control Board, Sacramento.

Foe, C., M. Stephenson, and S. Stanish. 2003. Pilot Transplant Studies with the Introduced Asiatic Clam, Corbicula fluminea, to Measure Methyl Mercury Accumulation in the Sacramento-San Joaquin Delta Estuary. In CALFED Final Report titled "An Assessment of Human Health and Ecological Impacts of Mercury in the Bay-Delta Watershed."

Furutani, A., and J. W. M. Rudd. 1980. Measurement of mercury methylation in lake water and sediment samples. Appl. Environ. Microbiol. 40:770-776.

Gill, G. A., N. S. Bloom, S. Cappellino, C. T. Driscoll, C. Dobbs, L. Mcshea, R. Mason, and J. W. M. Rudd. 1999. Sediment - water fluxes of mercury in Lavaca Bay, Texas. Environ. Sci. & Technol. 33:663-669.

Gill, G. A. 2003. Task 4B In CALFED Final Report titled "An Assessment of Human Health and Ecological Impacts of Mercury in the Bay-Delta Watershed."


Gilmour, C. C., E. A. Henry, and R. Mitchell. 1992. Sulfate stimulation of mercury methylation in freshwater sediments. Environ. Sci. Tech. 26:2281-2287.

Gilmour, C. C., G. S. Riedel, M. C. Ederington, J. T. Bell, J. M. Benoit, G. A. Gill, and M. C. Stordal. 1998. Methylmercury concentrations and production rates across a trophic gradient in the northern Everglades. Biogeochemistry 40:327-345.

Hurley, J. P., J. M. Benoit, C. L. Babiarz, M. M. Shafer, A. W. Andren, J. R. Sullivan, R. Hammond, and D. A. Webb. 1995. Influences of Watershed Characteristics on Mercury Levels in Wisconsin Rivers. Environ. Sci. & Technol. 29:1867-1875.

Hurley, J. P., D. P. Krabbenhoft, L. B. Cleckner, M. L. Olson, G. R. Aiken, and P. S. Rawlik. 1998. System controls on the aqueous distribution of mercury in the northern Florida Everglades. Biogeochemistry 40:293-311.

Kelly, C. A., J. W. M. Rudd, R. A. Bodaly, N. P. Roulet, V. L. St.Louis, A. Heyes, T. R. Moore, S. Schiff, R. Aravena, K. J. Scott, B. Dyck, R. Harris, B. Warner, and G. Edwards. 1997. Increases in fluxes of greenhouse gases and methyl mercury following flooding of an experimental reservoir. Environ. Sci. Technol. 31:1334-1344.

Kelly, C. A., J. W. M. Rudd, V. L. St. Louis, and A. Heyes. 1995. Is total mercury concentration a good predictor of methyl mercury concentration in aquatic systems. Water, Air, and Soil Pollution 80:715-724.

Krabbenhoft, D. P., J. G. Wiener, W. G. Brumbaugh, M. L. Olson, J. F. DeWild, and T. J. Sabin. 1999. A national pilot study of mercury contamination of aquatic ecostystems along multiple gradients. Pages 147-160 in D. W. Morganwalp and H. T. Buxton, editors. U.S. Geological Survey Toxic Substances Hydrology Program- Proceedings of the Technical Meeting, Charleston, South Carolina.

Larry Walker and Associates. 1997. Sacramento River Mercury Control Planning Project. Sacramento Regional County Sanitation District, Sacramento.

Lee, Y. H., and H. Hultberg. 1990. Methylmercury in some Swedish Surface Waters. Environ. Toxicol. Chem. 9:833-841.

Office of Environmental Health Hazard Assessment. 1994. Health Advisory on Catching and Eating Fish: Interim Sport Fish Advisory for San Francisco Bay. California Environmental Protection Agency, Sacramento.

Olson, B. H., and R. C. Cooper. 1976. Comparison of Aerobic and Anaerobic Methylation of Mercuric Chloride by San Francisco Bay Sediments. Water Research 10:113-116.

Nichols, F. H., J. E. Cloern, S. N. Luoma, and D. H. Peterson. 1986. The Modification of an Estuary. Science 231:567-573

Paquette, K., and G. Heltz. 1995. Solubility of Cinnabar (Red Hgs) and Implications for Mercury Speciation in Sulfidic Waters. Water, Air and Soil Pollution 80:1053-1056.

Puckett, H. M., and B. H. van Buuren. 2000. Quality Assurance Project Plan for the CALFED Project: "An Assessment of Ecological and Human Health Impacts of Mercury in the Bay-Delta Watershed". California Department of Fish and Game, Monterey.

Ravichandran, M., G. R. Aiken, M. M. Reddy, and J. Ryan. 1998. Enhanched Dissolution of Cinnabar (Mercury Sulfide) by Dissolved Organic Matter Isolated from the Florida Everglades. Environ. Sci. & Technol. 32:3305-3311.


Regnell, O., G. Ewald, and E. Lord. 1997. Factors controlling temporal variation in methyl mercury levels in sediment and water in a seasonally stratified lake. Limnol. Oceanogr. 42:1784-1795.

Roth, D. A., H. E. Taylor, J. L. Domagalski, P. D. Dileanis, D. B. Peart, R. C. Antweiler, and C. N. Alpers. 2001. Distribution of Inorganic Mercury in Sacramento River Water and Suspended Colloidal Sediment Material. Arch. Environ. Contam. Toxicol. 40:161-142.

Rudd, J. W. M. 1995. Sources of methylmercury to freshwater ecosystems. A review. Water, Air, Soil, Pollution 80:697-713.

Rudd, J. W. M., M. A. Turner, A. Furutani, A. Swick, and B. E. Townsend. 1983. The English-Wabigoon River system: 1. A synthesis of recent research with a view towards mercury amelioration. Can. J. Fish. Aquat. Sci. 40:2206-2217.

San Francisco Bay Regional Water Quality Control Board. 1995. Contaminant Levels in Fish Tissue from San Francisco Bay. San Francisco Regional Board, State Water Resources, and California Department of Fish and Game, San Francisco.

Slotten, D. G., S. M. Ayers, T. H. Suchanek, R. D. Weyand, A. M. Liston, C. Asher, D. C. Nelson, and B. Johnson. 2003. The Effects of Wetland Restoration on the Production and Bioaccumulation of Methylmercury in the Sacramento-San Joaquin Delta, California. In CALFED Final Report titled "An Assessment of Human Health and Ecological Impacts of Mercury in the Bay-Delta Watershed."

St. Louis, V. L., J. W. M. Rudd, C. A. Kelly, and L. A. Barrie. 1995. Wet Deposition of Methyl Mercury in Northwestern Ontario Compared to other Geographic Locations. Water, Air and Soil Pollution 80:405-414.

St. Louis, V. L., J. W. M. Rudd, C. A. Kelly, K. G. Beaty, N. S. Bloom, and R. J. Flett. 1994. Importance of wetlands as sources of methyl mercury to boreal forest ecosystems. Can. J. Fish. Aquat. Sci. 51:1065-1076.

van Buuren, B. H. 2003. The CALFED Mercury Project Quality Assurance Oversight Final Report. In CALFED Final Report titled " An Assessment of Human Health and Ecological Impacts of Mercury in the Bay-Delta Watershed."

Wallschlager, D., M. V. M. Desai, M. Spengler, and R. D. Wilken. 1998. Mercury Speciation in Floodplain Soils and Sediments Along a Contaminated River Transect. J. Environ. Qual. 27:1034-1044.

White, R. F., P. A. Grandjean, and P. Weihe. 1995. An Overview of Human Studies on CNS Effect of Methyl Mercury. In: National forum on Mercury in Fish. EPA 823-R-95-002, Environmental Protection Agency, Office of Water.

Wright, D. R., and R. D. Hamilton. 1982. Release of Methyl Mercury from Sediments: Effects of Mercury Concentration, Low Temperature, and Nutrient Addition. Can. J. Fish. Aquat. Sci. 39:1459-1466.


Table 1. A list of habitat types defined for the delta with the corresponding area, number of samples collected, and MMHg

concentration.

Habitat Type AreaKm2

Number of samples

[MMHg] (ng g-1)

Upland 3751 2 .19 ± .03

Riparian Woodland 41 4 .24 ± .08

Mudflats 3 2 .50 ± .32

Open Water 238 36 .54 ± .06

Seasonal Wetlands 167 3 .55 ± .06

Farmed Wetlands 1447 8 .71 ± .20

Lakes and Ponds 35 2 1.26 ± .43

Marsh 51 7 1.46 ± .35

Total 5733


Table 2. Name, location, and habitat type of seasonal monitoring sites

Name Latitude Longitude Habitat Classification

Prospect Slough 38.28961 121.66239 Open water / fringe wetland

Sherman Island 38.04120 121.81930 Tidal wetland / riparian marshland

Franks Tract 38.05250 121.59191 Open water

Connection S 38.01250 121.55002 Channeled waterway /fringe wetland

Cosumnes River 38.25823 121.42618 Riparian marshland

White Slough 38.08723 121.47837 Fringe tule


Table 3. Name, location, and habitat type of wetland sampling sites

Name Latitude Longitude Habitat Classification

Mandeville Cut 38.06152 121.53774 Peaty sediment, Tule wetland

Fourteen Mile Slough 38.00659 121.39224 Silty sediment, Wetland, backwater

Webber Point 38.00965 121.44651 Silty sediment, Wetland, backwater

Control 38.00397 121.44416 San Joaquin River Channel


Table 4. Relative percent difference(RPD) between field replicates for sediment samples collected October 1999 and analyzed for methyl mercury.

Station Number Station Name Rep 1 (ng/g)

Rep 2 (ng/g)

RPD

47 Sacramento River 0.18 0.30 50.3

54 Snodgrass River 0.26 0.08 108.3

91 SJR Jersey Pt. 0.04 0.06 48.8

95 San Joaquin River <MDL <MDL

104 Kimball Island 0.40 0.49 20.7

105 San Joaquin River 0.44 0.21 67.8

175 SJR @ Landers Ave 0.04 0.03 28.7

203 Port of Stockton 0.32 0.40 22.1

213 Liberty island 0.47 0.25 62.5

357 Cosumnes at Wilton 0.10 0.13 19.3

359 Cosumnes River 0.04 0.05 26.7

990 Cache Creek 0.13 0.21 48.7

991 Capay Dam 5.87 2.20 91.0


Table 4. Relative percent difference (RPD) between field replicates for sediment samples collected October 1999 and analyzed for total mercury.

Station Number Station Name Rep 1 (ng/g)

Rep 2 (ng/g)

RPD

36 Cache Slough 130.7 125.3 4.2

47 Sacramento River 61.1 32.6 60.7

50 Georgina Slough 81.1 85.1 4.8

54 Snodgrass River 46.1 27.6 50.3

64 Beaver Slough 190.1 160.3 17.0

67 Tower Park 135.7 156.3 14.1

71 Little Connection S 82.5 35.1 80.7

82 Connection Slough 110.9 125.8 12.6

85 Frank's Tract 78.5 68.2 14.0

88 Fisherman's Slough 156.8 206.9 27.5

91 SJR Jersey Pt. 33.1 79.9 82.9

93 Big Break 151.5 175.6 14.7


103 Antioch Dunes 57.6 100.4 54.2

104 Kimball Island 235.7 192.1 20.4


123 Suison Bay Marsh 178.1 250.5 33.8

175 SJR @ Landers <MDL <MDL

203 Port of Stockton 193.8 181.1 6.8

215 Liberty Cut Marsh 312.6 220.3 34.6

357 Cosumnes at Wilton <MDL <MDL

359 Cosumnes River <MDL <MDL

990 Cache Creek 93.0 86.6 7.1


Picture 1. A picture showing the fringe marsh habitat along rip-rap waterways at Mandeville Island located near Franks Tract in the central Bay-Delta. (Photo by W. Heim)


Picture 2. This picture of Tower Park marina in the Bay-Delta illustrates the subsidence of croplands and the necessity of levees for prevention of flooding. (Photo by W. Heim)


Picture 3. An assembled Sludge O Matic with attached shutter trigger line. The door is shown in the closed position. The attachment of the pulley system with plastic ties is clearly shown as is the handle configurations and aircraft pin. (Photo by W. Heim)


Picture 4. Close-up photo of the topmost 0.5 cm of sediment captured cleanly using the SOM. The trap door of the sampler is shown in the open position with sediment ready to be transferred into sample jars. (Photo by W. Heim)


Picture 5. Field team employing “clean hand techniques” at Sherman Island. Sediment is being sub-sampled from the SOM using a plastic scoop to transfer samples. “Clean hands” person is wearing clean room gloves during all sub-sampling. “Dirty hands” person never comes into contact with the sample but aids “clean hands” person as necessary. (Photo by R. Lehman)


Picture 6. Shallow water environment of East Columbia Cut characterized by a mixed sediment type of sand and silt with an abundance of floating and submerged aquatic vegetation.


Figure 1. Map showing the spatial survey sampling locations (○) within the delta and Grizzly Bay.


Figure 2. Total mercury surficial (0-0.5 cm) sediment concentrations of San Francisco Bay are indicated by purple scale bars with the key given in the lower left. The Key to Features shown in the upper right gives a color code for different habitat types found in the Bay as described by the National Wetlands Inventory. Sampling locations are represented as red circles.


Figure 3. Methyl mercury surficial (0-0.5 cm) sediment concentrations of San Francisco Bay are indicated by red scale bars with the key given in the lower left. The Key to Features shown in the upper right gives a color code for different habitat types found in the Bay as described by the National Wetlands Inventory. Sampling locations are represented as purple circles.


Figure 4. Methyl mercury to total mercury ratios (as a percent) in surficial (0-0.5 cm) sediment of San Francisco Bay are indicated by aqua scale bars with the key given in the lower left. The Key to Features shown in the upper right gives a color code for different habitat types found in the Bay as described by the National Wetlands Inventory. Sampling locations are represented as pink circles.


*

***

* *ProspectSlough

CosumnesRiver

ShermanIsland

Frank'sTract

ConnectionSlough

White Slough

GrizzlyBay

Suisun Bay

STOCKTO

N

MokelumneRiverSacramento River

San Joaquin River

Study Area

California

0

0

10 mile

10 20 km

Figure 5. A map of the San Francisco Bay-Delta showing the surrounding tributaries and Bays. The six seasonal sampling stations are shown with asterisks and the following names: Prospect Slough, Cosumnes River, White Slough, Franks Tract, Connection Slough, and Sherman Island. The dashed line circle represents the approximate boundary of an area within the study area operationally defined as the “central delta”.


Figure 6. Map showing the sampling sites for the mercury bioavailability study. The map key shows the symbol defining each sampling type.


Figure 7. Small scale variation in methyl mercury concentrations at East Columbia Cut in the central Bay-Delta. Field Replicates are shown with error bars as the standard deviation of analytical triplicates.


Sacramento River

San Joaquin River

Prospect Slough Cosumnes River

Franks Tract

Study Area

California 0

0

10 mile

10 20 km

GrizzlyBay

1.0

0.0

0.5

[THg](ppm dry wt sediment)

= below detection

Figure 8. Total mercury surficial (0-0.5 cm) sediment concentration for locations within the Delta and surrounding tributaries and bays. Mercury concentration for each location sampled is represented by a scale bar and the key is given in ppm dry weight sediment. Sampled locations with mercury values less than the method detection limit of 0.10 ppm are represented by solid black circles.


Figure 9. Correlation between total mercury concentration and longitude (R2 = 0.34) for samples collected from the Bay-Delta between 38.00° and 38.15° latitude.

0

100

200

300

400

500

600

121.2121.4121.6121.8122.0122.2122.4122.6

Longitude

Tota

l Mer

cury

(ppm

)


Sacramento River

San Joaquin River


Franks Tract

Study Area

California 0

0

10 mile

10 20 km

9.3 ppb

3.0 ppb

2.0

0.0

1.0

[MMHg](ppb dry wt sediment)

= below detection

Figure 10. Methyl mercury surficial (0-0.5 cm) sediment concentration for locations within the Delta and surrounding tributaries and bays. Methyl mercury concentration for each location sampled is represented by a scale bar and the key is given in ppb dry weight sediment. Two sites are shown with the methyl mercury concentrations explicitly expressed (in ppb) as the concentration of methyl mercury was large in comparison to all other samples. Sampled locations with methyl mercury values less than the method detection limit of 0.019 ppb are represented by solid black circles. The dashed line circle represents the approximate boundary of the study area operationally defined as the “central delta”.


Sacramento River

San Joaquin River


Franks Tract

Study Area

California 0

0

10 mile

10 20 km

5.0 %

0.0

2.5 %

MMHg percent of HgT= below detection

Figure 11. Methyl mercury to total mercury ratios (as a percent) in surficial (0-0.5 cm) sediment for locations within the Delta and surrounding tributaries and Bays. Grey scale bars are indicative of an upper limit ratio, as the methyl mercury to total mercury ratio was calculated using the method detection limit for total mercury. Solid black circles were used to indicate locations with methyl mercury values and total mercury values less than method detection limits. The dashed line circle represents the approximate boundary of the study area operationally defined as the “central delta”.


Figure 12. Correlation between methyl mercury and percent loss on ignition in Delta sediment samples collected during the winter of 1999 (R2 = 0.32).

0

1

2

3

0.0 1.0 2.0 3.0 4.0 5.0

% Loss on Ignition

Met

hyl M

ercu

ry (p

pb)


Figure 13. Correlation between methyl mercury and total mercury in Delta sediment samples collected during the winter of 1999 (R2 = 0.19). Circled triangles were considered as outliers and were left out of the correlation analysis.

0

1

2

3

4

0.0 100.0 200.0 300.0 400.0 500.0

Total Mercury (ppb)

Met

hyl M

ercu

ry (p

pb)


Figure 14. Seasonal changes in methyl mercury (▲), total mercury (■), and the methyl mercury to total mercury ratio (as a percent)(♦) in surficial sediment (0-0.5 cm) of the Bay-Delta. Methyl mercury (ppb) and methyl mercury to total mercury ratio (as %) share the primary y-axis. Total mercury is on the second y-axis (log scale). Error bars represent analytical uncertainty rather than field duplication.

0

2

4

6

8

10

M J J A S O N D J F M A M J J A S O N

MM

Hg

(ppb

)M

MH

g/H

gT

1

10

100

1000

2000 2001

a) Prospect Slough

0

2

4

6

8

10

M J J A S O N D J F M A M J J A S O N1

10

100

1000

Tota

l Mer

cury

(ppb

)

2000 2001

b) Sherman Island

0

2

4

6

8

10


MM

Hg

(ppb

)M

MH

g/H

gT

1

10

100

1000

2000 2001

e) White Slough

0

2

4

6

8

10


10

100

1000

Tota

l Mer

cury

(ppb

)

2000 2001

f) Cosumnes River

d) Connection Slough

0

2

4

6

8

10


10

100

1000

Tota

l Mer

cury

(ppb

)

2000 2001

0

2

4

6

8

10


MM

Hg

(ppb

)M

MH

g/H

gT

1

10

100

1000

2000 2001

c) Franks Tract

0

2

4

6

8

10


MM

Hg

(ppb

)M

MH

g/H

gT

1

10

100

1000

2000 2001

a) Prospect Slough

0

2

4

6

8

10


MM

Hg

(ppb

)M

MH

g/H

gT

1

10

100

1000

2000 2001

a) Prospect Slough

0

2

4

6

8

10


10

100

1000

Tota

l Mer

cury

(ppb

)

2000 2001

b) Sherman Island

0

2

4

6

8

10


10

100

1000

Tota

l Mer

cury

(ppb

)

2000 2001

b) Sherman Island

0

2

4

6

8

10


MM

Hg

(ppb

)M

MH

g/H

gT

1

10

100

1000

2000 2001

e) White Slough

0

2

4

6

8

10


MM

Hg

(ppb

)M

MH

g/H

gT

1

10

100

1000

2000 2001

e) White Slough

0

2

4

6

8

10


10

100

1000

Tota

l Mer

cury

(ppb

)

2000 2001

f) Cosumnes River

0

2

4

6

8

10


10

100

1000

Tota

l Mer

cury

(ppb

)

2000 2001

f) Cosumnes River


0

2

4

6

8

10


10

100

1000

Tota

l Mer

cury

(ppb

)

2000 2001


0

2

4

6

8

10


10

100

1000

Tota

l Mer

cury

(ppb

)

2000 2001

0

2

4

6

8

10


MM

Hg

(ppb

)M

MH

g/H

gT

1

10

100

1000

2000 2001

c) Franks Tract

0

2

4

6

8

10


MM

Hg

(ppb

)M

MH

g/H

gT

1

10

100

1000

2000 2001

c) Franks Tract


Figure 15. Methyl mercury and total mercury concentration of surficial sediment samples collected at Franks Tract (□), White Slough (∆), Connection Slough (x), Prospect Slough (○), Cosumnes River ( ), and Sherman Island (●). Red markers indicate samples collected during the seasonal peak of in situ methyl mercury production. A positive correlation between methyl mercury and total mercury is shown only for samples collected at Sherman Island (regression coefficient of R2 = 0.40).

0

2

4

6

8

10

0 50 100 150 200 250 300 350 400Total Mercury (ppb)

Met

hyl M

ercu

ry (p

pb)


Figure 16. Three panels showing total mercury concentration, methyl mercury concentration, and the methyl mercury to total mercury ratio (as a percent) in unsieved surficial sediments collected from coastal and Sierra lakes, mountain streams, and valley streams. Dark colored bars are from the coastal range and light colored bars are from the Sierra range. Each bar is the averaged value of many locations within a particular classification. The error bars are the standard error of samples within a particular classification.

1

10

100

1000

10000

100000

Lakes Mountain Streams Valley Streams

Tota

l Mer

cury

(ng/

g)

00.5

11.5

22.5

33.5


MM

Hg

(ng/

g)

0.00

1.00

2.00

3.00

4.00


MM

Hg/

HgT

(%)


Figure 17. Three panels showing total mercury concentration, methyl mercury concentration, and the methyl mercury to total mercury ratio (as a percent) in sieved surficial sediments collected from coastal and Sierra lakes, mountain streams, and valley streams. Dark colored bars are from the coastal range and light colored bars are from the Sierra range. Each bar is the averaged value of many locations within a particular classification. The error bars are the standard error of samples within a particular classification.

110

1001000

10000100000


Tota

l Mer

cury

(ng/

g)

0123456


MM

Hg

(ng/

g)

0

1

2

3

4


MM

Hg/

HgT

(%)


0

2

4

6

8

10

Mandeville Cut 14 mile Slough Webber Pt. Controls

MM

Hg

(ng/

g)

Figure 18. Methyl mercury concentrations in surficial sediments collected from three wetland areas and one control area outside the wetland area. Dark colored bars are from the interior of the wetlands, medium colored bars are mid way between the interior and fringe of the wetland, and the light colored bars are from the outer fringe of the wetland areas. The bars represent the average of field duplicates and the error bars are the range of field duplicates.


0

1

2

3

4

Mandeville Cut 14 mile Slough Webber Point Control

MM

Hg/

HgT

(%)

Figure 19. Methyl mercury to total mercury ratio as a percent in surficial sediments collected from three wetland areas and one control area outside the wetland area. Dark colored bars are from the interior of the wetlands, medium colored bars are mid way between the interior and fringe of the wetland, and light colored bars are from the outer fringe of the wetland areas. The bars represent the average of field duplicates and the error bars are the range of field duplicates.

methyl and total mercury spatial and temporal …...2001). as a result, the fish, water, and...

Documents